The Testosterone Trial is a randomized, placebo-controlled group of six closely coordinated trials in 800 men >65 years old who have unequivocally low serum testosterone of the effect of testosterone treatment for one year on physical function, sexual function, vitality, cognition, anemia and cardiovascular risk. The Testosterone Trial has been funded and is expected to begin enrolling subjects in the summer of 2010. We propose adding a Bone Trial in a subset of 200 men who enroll in The Testosterone Trial. The rationale for The Bone Trial is that although the evidence is clear that testosterone greatly improves trabecular bone density and strength in men who are severely hypogonadal due to known pituitary or testicular disease, the evidence that testosterone improves bone in elderly men who have more modest degrees of testosterone deficiency is equivocal. We propose taking advantage of the availability of men in The Testosterone Trial who have been recruited and screened and found to have unequivocally low serum testosterone concentrations and have agreed to participate in this trial to determine definitively if testosterone treatment will improve their bone quality. The overall specific aim of this trial is to test the hypothesis that testosterone treatment for one year will improve bone quality more than placebo in men >65 years who have serum testosterone concentrations <275 ng/dL. The primary specific end point is volumetric bone mineral density (vBMD) of the spine by quantitated computed tomography (QCT). Secondary end points are vBMD of the hip; bone strength and strength-to-density ratio in the spine and hip by finite element analysis determined from the QCT data; areal BMD of the lumbar spine, proximal hip and distal radius by DXA; serum markers of bone turnover; and clinical fractures. Subjects (200) will be recruited from among men who have qualified for one of the trials in The Testosterone Trial and who have no diseases and are taking no medications known to affect bone. They will be randomized to receive testosterone or placebo gel double blindly for one year. QCT and DXA of the spine, hip and distal radius will be performed at baseline and 12 months. Bone markers will be determined at baseline, 1, 2, 3, 6, and 12 months. Clinical fractures will be assessed at 3, 6, 9, 12, 18 and 24 months. The significance of The Bone Trial is that it would be the first study to assess the effect of testosterone on volumetric BMD in men >65 with low testosterone and, more importantly, on bone strength. Demonstration that testosterone improves bone strength in these men would provide the rationale for a larger, longer trial to determine if testosterone treatment reduces clinical fractures as well. The Bone Trial will also provide data on clinical fractures that will allow sample size estimation for a larger, longer trial.